Deliberate Practice and Proposed Limits on the Effects of Practice on the Acquisition of Expert Performance: Why the Original Definition Matters and Recommendations for Future Research

Abstract

Over 25 years ago Ericsson et al. (1993) published the results of their search for the most effective forms of training in music, a domain where knowledge of effective training has been accumulated over centuries. At music academies master teachers provide students individualized instruction and help them identify goals and methods for their practice sessions between meetings - this form of solitary practice was named deliberate practice, and its accumulated duration during development was found to distinguish groups with differing levels of attained music performance. In an influential meta-analysis Macnamara et al. (2014) identified studies that had collected estimates of practice accumulated during development and attained performance and reported that individual differences in deliberate practice accounted for only 14% of variance in performance. Their definition of "deliberate practice" differs significantly from the original definition of deliberate practice and will henceforth be referred to as structured practice. We explicate three criteria for reproducible performance and purposeful/deliberate practice and exclude all effect sizes considered by Macnamara et al. (2014) that were based on data not meeting these criteria. A reanalysis of the remaining effects estimated that accumulated duration of practice explained considerably more variance in performance (29 and 61% after attenuation correction). We also address the argument that the limited amount of variance explained by the duration of practice necessarily implies an important role of genetic factors, and we report that genetic effects have so far accounted for remarkably small amounts of variance - with exception of genetic influences of height and body size. The paper concludes with recommendations for how future research on purposeful and deliberate practice can go beyond recording only the duration of practice to measuring the quality of practice involving concentration, analysis, and problem solving to identify conditions for the most effective forms of training.

Keywords: deliberate practice; expert performance; heritability; mental representation; practice effects.

FIGURE 1

Three types of systems of mental representations that mediate expert music performance and its continued improvement during deliberate practice. Adapted from Figure 6 in Ericsson (1998).

FIGURE 2

Pianists’ retrospective estimates of their weekly practice as a function of age. Data in the left panel are aggregated for young and older pianists. Data points above the minimum ages (20 for the young pianists and 52 for the older pianists) include at least 50% of the participants in each group (reproduced with permission of Figure 7 in Krampe and Ericsson, 1996).

FIGURE 3

Flow diagram of applying revised inclusionary criteria to estimates of the effects of deliberate practice on performance considered by Macnamara et al. (2014).

FIGURE 4

Correlations between purposeful or deliberate practice and performance. Squares represent correlation coefficients for the effects of practice on performance for each listed sample, and lines represent 95% confidence intervals. The marker at the bottom shows the weighted mean correlation. Study naming conventions were kept consistent with those used by Macnamara et al. (2014) for comparison purposes, indicating separate samples (e.g., S1) or measures of practice (e.g., DP2) or performance (e.g., M1).